Glycoprotein biosynthesis in animal cells grown in suspension culture. Assembly of lipid-linked saccharides and formation of protein-bound ''high-mannose'' oligosaccharides.

Glycoprotein biosynthesis was studied with mouse L-cells grown in suspension culture. Glucose-deprived cells incorporated [3H]mannose into 'high-mannose' protein-bound oligosaccharides and a few relatively high-molecular-weight lipid-linked oligosaccharides. The latter were retained by DEAE-cellulose and turned over quite slowly during pulse--chase experiments. Increased heterogeneity in size of lipid-linked oligosaccharides developed during prolonged glucose deprivation. Sequential elongation of lipid-linked oligosaccharides was also observed, and conditions that prevented the assembly of the higher lipid-linked oligosaccharides also prevented the formation of the larger protein-bound 'high-mannose' oligosaccharides. In parallel experiments, [3H]mannose was incorporated into a total polyribosome fraction, suggesting that mannose residues were transferred co-translationally to nascent protein. Membrane preparations from these cells catalysed the assembly from UDP-N-acetyl-D-[6-3H]glucosamine and GDP-D-[U-14C]mannose of polyisoprenyl diphosphate derivatives whose oligosaccharide moieties were heterogeneous in size. Elongation of the N-acetyl-D-[6-3H]glucosamine-initiated glycolipids with mannose residues produced several higher lipid-linked oligosaccharides similar to those seen during glucose deprivation in vivo. Glucosylation of these mannose-containing oligosaccharides from UDP-D-[6-3H]glucose was restricted to those of a relatively high molecular weight. Protein-bound saccharides formed in vitro were mainly smaller in size than those assembled on the lipid acceptors. These results support the involvement of lipid-linked saccharides in the synthesis of asparagine-linked glycoproteins, but show both in vivo and in vitro that protein-bound 'high-mannose' oligosaccharide formation can occur independently of higher lipid-linked oligosaccharide synthesis.     

Fluoroglucose-inhibition of protein glycosylation in vivo. Inhibition of mannose and glucose incorporation into lipid-linked oligosaccharides

The effects of the glycosylation inhibitor 2-deoxy-2-fluoro-D-glucose on the formation of the lipid-linked oligosaccharides and monosaccharides that are involved in protein glycosylation were investigated. In chick embryo cells treated with fluoroglucose the formation of lipid-linked oligosaccharides cannot go to completion and oligosaccharides with decreased amounts of glucose and mannose can be detected. These oligosaccharides are probably biosynthetic intermediates and serve as acceptors of sugar residues while reversing fluoroglucose-inhibition by the addition of mannose and glucose to the culture medium. In contrast to deoxyglucose, fluoroglucose was not incorporated into lipid-linked oligosaccharides. Fluoroglucose inhibits the formation in vivo of dolichyl phosphate glucose and dolichyl phosphate mannose, but not the transfer of those sugar residues from the lipid monophosphate derivative to the lipid-linked oligosaccharides. The pool size of UDP-glucose, but not of GDP-mannose and UDP-N-acetylglucosamine, was decreased. Also, the formation of lipid-linked N-acetylglucosamine was not affected by fluoroglucose. Fluoroglucose was applied to deplete cellular membranes of endogenous lipid-linked mannose and glucose, and can possibly be used to discern different pathways of glycosylation.     

Inhibition of beta-1,4-Glucan Biosynthesis by Deoxyglucose : The Effect on the Glucosylation of Lipid Intermediates

GDP- and UDP-deoxyglucose inhibit the incorporation of glucose from UDP-glucose into dolichyl phosphate glucose and dolichyl pyrophosphate oligosaccharides. GDP-deoxyglucose inhibits by competing with the physiological nucleotide sugars for dolichyl phosphate, and dolichyl phosphate deoxyglucose is formed. This inhibition is reversed by excess of dolichyl phosphate. UDP-deoxyglucose does not give rise to a lipid-linked derivative, and inhibition by this analog is not reversed by dolichyl phosphate. The UDP- and GDP-derivatives of deoxyglucose inhibit the incorporation of glucose into glucose-containing glycoproteins. This effect seems to be the result of the inhibition of lipid intermediates glucosylation and is comparable to the effect produced by coumarin. Cellulose synthetase activity is not affected by UDP- or GDP-deoxyglucose. On the other hand, deoxyglucose inhibits the formation of β-1,4-glucans in vivo.     

Dolichol-dependent protein glycosylation in a deoxyglucose- resistant baby-hamster-kidney cell line

The formation in vivo of lipid-linked oligosaccharides is inhibited by deoxlucose in wild-type BHK cells but not in a cell-line (dG_R) selected for resistance towards deoxyglucose. On the other hand, the formation in vitro of lipid-linked oligosaccharides by membranes from dG_R (and wild-type) cells is inhibited by GDPddeoxyglucose, the main metabolite responsible for inhibition of protein glycosylation by deoxyglucose. Our results suggest increased pools of GDP mannose and decreased amounts of GDPdeoxyglucose in the mutant cell line. The enlarged ratio of GDPmannose to GDPdeoxyglucos in the dG_R cells treated with deoxyglucose is shown to moderate the inhibition of formation of lipid-linked oligosaccharides, and this explains the capacity of the dG^R-cells grow in the presence of deoxyglucose.     

Inhibition of protein synthesis also inhibits synthesis of lipid-linked oligosaccharides.

Studies were initiated to determine whether the formation of lipid-linked oligosaccharides was coupled to the synthesis of protein. Canine kidney cells were grown with [2-3H]mannose or [3H]leucine in the presence of cycloheximide or puromycin and the effect of these inhibitors on the synthesis of proteins and lipid-linked oligosaccharides was measured. In all cases, the inhibition of protein synthesis resulted in a substantial inhibition in the incorporation of mannose into the lipid-linked oligosaccharides, although the synthesis of mannosyl-phosphoryl-dolichol was only slightly inhibited. Cycloheximide had no effect on the in vitro incorporation of mannose into lipid-linked oligosaccharides when GDP-[14C]mannose was incubated with aorta microsomal preparations. The inhibition of lipid-linked oligosaccharides was apparently not due to a decrease in the amount of glycosyltransferases as a result of protein degradation in the absence of protein synthesis, nor was it the result of a more rapid degradation of lipid-linked oligosaccharides. The inhibition also did not appear to be due to limitations in the available dolichyl-phosphate. The results suggest that the formation of lipid-linked oligosaccharides may be regulated by end product inhibition.     

Involvement of Lipid-linked Oligosaccharides in Synthesis of Storage Glycoproteins in Soybean Seeds

Membrane preparations from developing soybean (var. Prize) cotyledon tissue, at the time of synthesis of storage glycoproteins, catalyze the sequential assembly of lipid-linked oligosaccharides from uridine-5'-diphospho-N-acetyl-d-[6-(3)H] glucosamine and guanosine-5'diphospho-d-[U-(14)C]mannose. The maximum size of lipid-linked oligosaccharide that accumulates contains the equivalent of 10 saccharide units on the basis of Bio-Gel P-2 gel filtration studies. These lipid-linked oligosaccharides show similar characteristics to polyisoprenyl diphosphate derivatives on diethylaminoethyl-cellulose chromatography and are potential intermediates in glycoprotein biosynthesis in this tissue. These glycolipids do not appear to turn over in pulse-chase experiments and no completed storage glycoproteins were detected among the products of these incubations.Tissue slices from cotyledons at the same stage of development synthesize lipid-linked oligosaccharides from [(3)H]mannose and [(3)H]glucosamine with sizes equivalent to 1, 7, 10, and approximately 15 saccharide units. In pulse-chase experiments, the lipid-linked saccharides with the equivalent of 1 and 10 units rapidly turnover, whereas those with 7 and 15 units do not. Examination of the higher oligosaccharide peaks (10 and 15) by Bio-Gel P-4 gel filtration shows them to comprise 2 distinct subsets of oligosaccharides containing different proportions of glucosamine and mannose units. Tissue slices synthesize products which resemble the completed 7S storage glycoproteins as judged by similarity of molecular weight and precipitation with specific antisera. Analysis of the oligosaccharides obtained by hydrazinolysis of glycoproteins shows the presence of a similar size "high-mannose" type N-linked oligosaccharides as in other glycoproteins from animal and plant cells.     

Deoxyglucose inhibition of protein glycosylation: Effects of nucleotide deoxysugars on the formation of glucosylated lipid intermediates

The UDP-derivative of deoxyglucose (UDP-deoxyglucose) inhibits the formation of dolichyl monophosphate glucose (Dol-P-glucose) in chick embryo cell membranes but has no effect on Dol-PP di-N-acetylchitobiose [Dol-PP-(GlcNAc)₂]formation. The effects of UDP-deoxyglucose are not reversed by Dol-P, nor is Dol-P-deoxyglucose formed from this derivative. GDP-deoxyglucose inhibits formation of both Dol-P-glucose and Dol-PP-(GlcNAc)₂. It is shown that GDP-deoxyglucose inhibits in these cases by competition with physiological nucleotide sugars for Dol-P. GDP-deoxyglucose and UDP-deoxyglucose also prevent the attachment of the peripheral glucose residues in Dol-PP-(GlcNAc)₂-Mans_yGlc₃, the immediate precursor of protein-bound oligosaccharides. The inhibition by GDP-deoxyglucose is only in part reversed by Dol-P, probably because deoxyglucose is incorporated into the lipid-linked oligosaccharide instead of glucose.     

Incorporation of glucose into lipid-linked saccharides in aorta and its inhibition by amphomycin

The particulate enzyme from pig aorta catalyzed the transfer of glucose from UDP-glucose into glucosyl-phosphoryl-dolichol, into lipid-linked oligosaccharides, and into glycoprotein. Radioactive lipid-linked oligosaccharides were prepared by incubating the extracts with GDP-[¹⁴C]mannose and UDP-[³H]glucose. When the labeled oligosaccharides were run on Bio-Gel P-4, the two different labels did not exactly coincide; the ³H peak eluted slightly earlier indicating that it was of higher molecular weight than the ¹⁴C material, but there was considerable overlap. The purified oligosaccharide(s) contained glucose, mannose, and N-acetylglucosamine but the ratios of these sugars varied from one enzyme preparation to another, probably depending on the endogenous oligosaccaride-lipids present in the microsomal preparation. Treatment of the [³H]glucose-labeled oligosaccharide with α-mannosidase gave rise to a ³H-labeled oligosaccharide which moved somewhat faster on Bio-Gel P-4 than the original oligosaccharide, suggesting it had lost one or two sugar residues. These data indicate that mannose and glucose are in the same oligosaccharide. The antibiotic, amphomycin, inhibited the transfer of glucose from UDP-glucose into the lipid-linked saccharides. However the synthesis of glucosyl-phosphoryl-dolichol was much more sensitive then was the synthesis of lipid-linked oligosaccharides. The glucose-labeled oligosaccharide produced in the absence of amphomycin was of high molecular weight based on paper chromatography. But in the presence of partially inhibitory concentrations of antibiotic, the oligosaccharide migrated more rapidly on paper chromatograms. However, amphomycin had no effect on the synthesis of glucosyl-ceramide by the aorta extracts. In fact, the antibiotic may stimulate glucosyl-ceramide by making more of the substrate, UDP-glucose, available for synthesis of this lipid.     

Enzymatic transfer of mannose from mannosyl-phosphoryl-polyprenol to lipid-linked oligosaccharides by pig aorta.

A particulate enzyme preparation prepared from the intimal layer of pig aorta catalyzed the transfer of mannose from mannosyl-phosphoryl-polyprenol (MPP) into a series of oligosaccharides that were linked to lipid. The reaction required detergent with Triton X-100 and NP-40 being best at a concentration of 0.5%. Several other detergents were inactive or only slightly active. The pH optima for this activity was about 7 to 7.5 in Tris buffer and the apparent Km for MPP was about 2 x 10(-7) M. The reaction was not stimulated by the addition of divalent cation and, in fact, was inhibited by the high concentrations of cation. The addition of EDTA did not inhibit the transfer of mannose from MPP and was somewhat stimulatory. The transferase(s) activity was "solubilized" from the particles by treatment with Triton X-100. This solubilized enzyme still formed a series of lipid-linked oligosaccharides from either MPP or GDP-mannose. The oligosaccharides were released from the lipid by mild acid hydrolysis and were separated by paper chromatography. Some five or six radioactive oligosaccharides were formed from either MPP or from GDP-mannose and these oligosaccharides had similar mobilities upon paper chromatography. However, MPP was a better donor for the larger oligosaccharides (i.e. those containing 8, 9, or 10 sugar residues), whereas GDP-mannose was better for formation of the oligosaccharide containing 7 sugar residues. In the presence of EDTA and detergent no MPP was formed from GDP-mannose, but radioactivity was still incorporated into the lipid-linked oligosaccharides. Under these conditions essentially all of the radioactivity was in the oligosaccharide containing 7 sugar residues. Since much of this activity could be released as mannose by acetolysis, GDP-mannose may be the direct mannosyl donor for formation of 1 leads to 6 branches. Oligosaccharides 7, 8, 9, and 10 were isolated and partially characterized in terms of their molecular weights, sugar composition, susceptibility to alpha-mannosidase, and 14C products formed by acetolysis and periodate oxidation. The molecular weights ranged from 1310 for oligosaccharide 7 to 1750 for oligosaccharide 10. Hydrolysis of each oligosaccharide and reduction with NaB3H4 gave the expected ratio of [3H]hexitol to [3H]hexosaminitol based on the molecular weight of the oligosaccharide. However, the hexitol fraction contained [3H]mannitol and [3H]glucitol. Since the amount of radioactivity in glucitol was 2 to 4 times that in mannitol and since only glucosaminitol was found in the amino sugar peak, it seems likely that each 14C-oligosaccharide was contaminated with an unlabeled oligosaccharide of equal molecular weight containing glucose and GlcNAc. Acetolysis of the 14C-oligosaccharides gave rise to 14C peaks of mannose, mannobiose, and mannotriose. In the larger oligosaccharides, most of the radioactivity was in mannobiose whereas in oligosaccharide 7 most of the radioactivity was in mannose...     

Carbohydrate Composition of Variant-Specific Surface Antigen Glycoproteins from Trypanosoma brucei

SYNOPSIS. The carbohydrate of variant-specific surface antigen glycoproteins from bloodstream forms of 13 cloned variants of Trypanosoma brucei was analyzed by gas-liquid chromatography. The glycoproteins contained from 6 to 17% carbohydrate by weight, and all contained the same 4 sugars: mannose, galactose, glucose, and glucosamine (probably as N-acetyl-glucosamine). The glycoprotein from variant 048, strain 427 contained (±20%) 11 mannose, 4 galactose, 4 glucose, and 5 glucosamine residues/mole of glycoprotein (molecular weight 65,000). Glucose was an integral component of the glycoproteins, not dissociable by sodium dodecyl sulphate, 8 M urea, or 1 M acetic acid. Some of the glucose was dissociated by trichloroacetic acid. Most of the glycoproteins formed precipitin bands with concanavalin A in Ouchterlony double diffusion, but none formed such bands with wheat germ agglutinin or Ricinus communis lectin (molecular weight 120,000).     

Carbohydrate composition of variant-specific surface antigen glycoproteins from Trypanosoma brucei

The carbohydrate of variant-specific surface antigen glycoproteins from bloodstream forms of 13 cloned variants of Trypanosoma brucei was analyzed by gas-liquid chromatography. The glycoproteins contained from 6 to 17% carbohydrate by weight, and all contained the same 4 sugars: mannose, galactose, glucose, and glucosamine (probably as N-acetylglucosamine). The glycoprotein from variant 048, strain 427 contained (+20%) 11 mannose, 4 galactose, 4 glucose, and 5 glucosamine residues/mole of glycoprotein (molecular weight 65,000). Glucose was an intergral component of the glycoproteins, not dissociable by sodium dodecyl sulphate, 8 M urea, or 1 M acetic acid. Some of the glucose was dissociated by trichloroacetic acid. Most of the glycoproteins formed precipitin with concanavalin A in Ouchterlony double diffusion, but none formed such bands with wheat germ agglutinin or Ricinus communis lectin (molecular weight 120, 000).     

A Mucor pusillus mutant defective in asparagine-linked glycosylation.

A Mucor pusillus mutant defective in asparagine-linked glycosylation was found in our stock cultures. This mutant, designated 1116, secreted aspartic proteinase (MPP) in a less-glycosylated form than that secreted by the wild-type strain. Analysis of enzyme susceptibility, lectin binding, and carbohydrate composition indicated that this mutant secreted three glycoforms of MPPs, one of which contained no carbohydrate; the other two had truncated asparagine-linked oligosaccharide chains such as Man0-1GlcNAc2. Further analysis using oligosaccharide processing inhibitors, such as castanospermine, 1-deoxynojirimycin and N-methyldeoxynojirimycin, suggested that MPPs in the mutant were glycosylated through a transfer of the truncated lipid-linked oligosaccharides, Man0-1GlcNAc2, to the MPP protein but not through an aberrant processing. In addition, genetic studies with forced primary heterokaryons indicated that the mutation in strain 1116 was recessive.     

An inhibitor of mannosylation of retinyl-phosphate

The guanosine disphospate and uridine diphosphate esters of the antiviral sugar analog 2-deoxy-2-fluoro-D-glucose (GDP-FGlc and UDP-FGlc) were synthesized and tested as inhibitors of formation of lipid-linked sugars in cell-free extracts . Formation of dolichol-phosphate mannose and of dolichol-diphosphate di-N-acetylchitobiose were not inhibited by either sugar nucleotide. Formation of dolichol-phosphate glucose was inhibited by UDP-FGlc, not by GDP-FGlc. Although GDP-FGlc did not inhibit formation of dollchol-phosphate mannose, it did inhibit formation of retinol-phosphate mannose from retinol-phosphate and GDP-Man. This inhibition was not reversed by exogenous retinol-phosphate, nor was FGIc from GDP-FGlc incorporated into retinolphosphate-linked derivatives. As FGLc inhibits formation of dolichol-phosphate mannose in intact cells, but does not decrease pool sizes of GDP-Man and dolichol-phosphate (Datema et al., 1980, Eur. J. Biochem.109, 331–341), we discuss that inhibition of formation of retinol-phosphate mannose by one of the metabolites of FGlc, namely GDP-FGlc, may lead to decreased synthesis of dolichol-phosphate mannose in FGlc-treated cells. This implies a role for vitamin A in the dolichol cycle of protein glycosylation.     

The effect of mannosamine on the formation of lipid-linked oligosaccharides and glycoproteins in canine kidney cells

Madin-Darby canine kidney (MDCK) cells normally form lipid-linked oligosaccharides having mostly the Glc3Man9GlcNAc2 oligosaccharide. However, when MDCK cells are incubated in 1 to 10 mM mannosamine and labeled with [2-3H]mannose, the major oligosaccharides associated with the dolichol were Man5GlcNAc2 and Man6GlcNAc2 structures. Since both of these oligosaccharides were susceptible to digestion by endo-beta-N-acetylglucosaminidase H, the Man5GlcNAc2 must be different in structure than the Man5GlcNAc2 usually found as a biosynthetic intermediate in the lipid-linked oligosaccharides. Methylation analysis also indicated that this Man5GlcNAc2 contained 1----3 linked mannose residues. Since pulse chase studies indicated that the lesion was in biosynthesis, it appears that mannosamine inhibits the in vivo formation of lipid-linked oligosaccharides perhaps by inhibiting the alpha-1,2-mannosyl transferases. Although the lipid-linked oligosaccharides produced in the presence of mannosamine were smaller in size than those of control cells and did not contain glucose, the oligosaccharides were still transferred in vivo to protein. Furthermore, the oligosaccharide portions of the glycoproteins were still processed as shown by the fact that the glycopeptides were of the complex and hybrid types and were labeled with [3H]mannose or [3H]galactose. In contrast, control cells produced complex and high-mannose structures but no hybrid oligosaccharides were detected. The inhibition by mannosamine could be overcome by adding high concentrations of glucose to the medium.     

Studies on the structure of lipopolysaccharides of Salmonella minnesotta and Salmonella typhimurium R strains

1. The composition of the lipopolysaccharides and the corresponding lipid-free polysaccharides from four R-mutants of Salmonella has been studied. All the lipopolysaccharides, from RI and RII serotypes contained d-glucose, d-galactose, heptose, N-acetylglucosamine and 3-deoxy-2-oxo-octonate. The polysaccharide obtained from the RII lipopolysaccharides also contained all these sugars. The polysaccharides from RI lipopolysaccharides lacked N-acetylglucosamine. 2. From partial hydrolysates of the lipopolysaccharides, a number of oligosaccharides have been isolated and partially characterized. Oligosaccharides containing N-acetylglucosamine or glucosamine were obtained only from RII lipopolysaccharides. Several oligosaccharides composed of glucose and galactose were common to RI and RII preparations. 3. A structural unit, based on the oligosaccharides found, is proposed for the RII lipopolysaccharide. It contains the sequence: alpha-N-acetylglucosaminyl- alpha-glucosyl-alpha-galactosyl-glucosyl.... A second alpha-galactosyl residue is bound to position 6 of the last glucosyl group. The complete unit is believed to to be attached to a polyheptose phosphate backbone in the RII antigen. 4. The RI lipopolysaccharide of Salmonella minnesota contains an analogous structure lacking the terminal N-acetylglucosamine residue. 5. A basal structure common to the lipopolysaccharides of several Salmonella species is proposed.     

Regulation of mammary gland metabolism: pathways of glucose utilization, metabolite profile and hormone response of a modified mammary gland cell preparation

Crude membrane preparations from chick embryo cells catalyse the formation of dolichyl-di-N-acetylchitobiosyl diphosphate [Dol-PP-(GlcNAc)2] from uridine diphosphate N-acetylglucosamine (UDP-GlcNAc). The formation of this glycolipid was stimulated by exogenous dolichyl phosphate and inhibited by tunicamycin. Adding GDP-mannose to the cell-free system containing Dol-PP-(GlcNAc)2 by preincubation led to the formation of a lipid-linked oligosaccharide, containing 8--9 sugar residues. The formation of lipid-linked oligosaccharides was inhibited by GDP-2-deoxy-D-glucose (GDP-dGlc): in this case Dol-PP-(Glc-NAc)2-dGlc accumulated. Subsequent additions of mannosyl residues to this trisaccharide-lipid to form lipid-linked oligosaccharides were not possible. Concomitantly the glycosylation of proteins was blocked. Partially inhibitory conditions were obtained by adding both GDP-dGlc and GDP-Man with an excess of GDP-dGlc. Glycosylation of proteins was observed but the glycopeptides did not contain 2-deoxyglucosyl residues. Also in these cases 2-deoxyglucose-containing glycolipids accumulated. The main glycolipid formed under these conditions was Dol-PP-(GlcNAc)2-Man-dGlc. Lipid-linked oligosaccharides containing 2-deoxyglucose were formed under these conditions, although in small amounts, but were not transferred to protein. So the molecular basis of the inhibitory action of 2-deoxyglucose on glycosylation of protein is the incorporation of 2-deoxyglucosyl residues during early phases of the biosynthesis of the lipid-linked oligosaccharides.     

Purification and characterization of a glycosylated proline-rich protein from human parotid saliva.

1. A glycosylated proline-rich protein (GPRP) was purified to homogeneity by subjecting parotid saliva to immunoaffinity, cation exchange, affinity and hydrophobic chromatography. 2. The purified GPRP had a molecular weight of 78 kDa as analyzed by SDS-PAGE. 3. The amino acid analysis revealed a preponderance of proline, glycine and glutamic acid/glutamine, which accounted for 77% of the total amino acids. 4. Cysteine, tyrosine or phenylalanine residues were not detected. 5. The glycoprotein contained 34% neutral sugars and the oligosaccharides were rich in mannose and N-acetylglucosamine, indicating that N-linked oligosaccharides were the predominant type of oligosaccharides in the molecule. 6. These observations were confirmed by treatment of the purified glycoprotein with specific N-glycosidase which removed the N-linked oligosaccharides leaving a core protein with an apparent molecular weight of 51 kDa. 7. The isoelectric point of GPRP was approx 7.0 and the molecule was not affected by reduction with 2-mercaptoethanol, indicating that no disulfide linkages were present. 8. The GPRP bound to hydroxyapatite and this binding could be partially inhibited by preincubation of the hydroxyapatite with parotid or submandibular saliva. 9. The purified GPRP also bound to a protein with an apparent molecular weight of 95 kDa present in submandibular saliva.     

Inhibition of lipid-linked saccharide synthesis by bacitracin.

The antibiotic bacitracin was found to inhibit the incorporation of mannose and GlcNAc from their respective sugar nucleotides into lipid-linked saccharides. The inhibition of both systems was apparent in the aorta particulate enzyme system but it was much more pronounced with the solubilized enzyme system. In both cases, GlcNAc incorporation into Dol-P-P-GlcNAc was more sensitive than mannose incorporation into Dol-P-Man, with 50% inhibition being seen at about 0.1–0.2 mm antibiotic. Bacitracin inhibition of mannose incorporation appeared to be overcome at high concentrations of dolichyl phosphate but, in these cases, an unexplained stimulation was observed. However, GlcNAc inhibition could not be overcome by high concentrations of dolichol phosphate, metal ion, or both together. Thus, the mechanism of inhibition by bacitracin is not clear. Bacitracin also inhibited the transfer of mannose from GDP-mannose to lipid-linked oligosaccharides and to glycoprotein in the particulate enzyme, as well as the transfer of radioactivity from Dol-P-Man or from lipid-linked oligosaccharides to glycoprotein. Thus, bacitracin apparently blocks each of the steps in the lipid-linked pathway. In yeast spheroplasts, bacitracin inhibited the incorporation of [¹⁴C]mannose into Dol-P-Man, into lipid-linked oligosaccharides, and into glycoprotein. However, in this case, the antibiotic also blocked the incorporation of leucine into protein. Bacitracin also inhibited the cell-free synthesis of mannosyl-phosphoryl-decaprenol in Mycobacterium smegmatis with 50% inhibition being observed at a concentration of about 0.5 mm.     

Carbohydrates of Influenza Virus. I. Glycopeptides Derived from Viral Glycoproteins After Labeling with Radioactive Sugars

The carbohydrate moiety of the influenza glycoproteins NA, HA₁, and HA₂ were analyzed by labeling with radioactive sugars. Analysis of glycopeptides obtained after digestion with Pronase indicated that there are at least two different types of carbohydrate side chains. The side chain of type I is composed of glucosamine, mannose, galactose, and fucose. It is found on NA, HA₁, and HA₂. The side chain of type II contains a high amount of mannose and is found only on NA and HA₂. The molecular weights of the corresponding glycopeptides obtained from virus grown in chicken embryo cells are 2,600 for type I and 2,000 for type II. The glycoproteins of virus grown in MDBK cells have a higher molecular weight than those of virus grown in chicken embryo cells, and there is a corresponding difference in the molecular weights of the glycopeptides. Under conditions of partial inhibition of glycosylation, virus particles were isolated that contained hemagglutinin with reduced carbohydrate content. Glycopeptide analysis indicated that this reduction is due to the lack of whole carbohydrate side chains and not to the incorporation of incomplete ones. This observation suggests that glycosylation of the viral glycoproteins involves en bloc transfer of the core sugars to the polypeptide chains.     

The effect of glycoprotein-processing inhibitors on the secretion of glycoproteins by Madin-Darby canine kidney cells

The effects of various glycoprotein-processing inhibitors on the biosynthesis and secretion of N-linked glycoproteins was examined in cultured Madin-Darby canine kidney (MDCK) cells. Since incorporation of [2-3H]mannose into lipid-linked saccharides and into glycoproteins was much greater in phosphate-buffered saline (PBS) than in serum-supplemented basal medium (BME), most experiments were done in PBS. Castanospermine, an inhibitor of glucosidase I, caused the formation of glycoproteins having mostly Glc3Man7-9(GlcNAc)2 structures; deoxymannojirimycin, an inhibitor of mannosidase I, gave mostly glycoproteins with Man9(GlcNAc)2 structures; swainsonine, an inhibitor of mannosidase II, caused the accumulation of hybrid types of oligosaccharides. Castanospermine and swainsonine, either in PBS or in BME medium, had no effect on the incorporation of [2-3H]mannose or [5,6-3H]leucine into the secreted glycoproteins and, in fact, there was some increase in mannose incorporation in their presence. These inhibitors also did not affect mannose incorporation into cellular glycoproteins nor did they affect the biosynthesis as measured by mannose incorporation into lipid-linked saccharides. On the other hand in PBS medium, deoxymannojirimycin, at 25 micrograms/mL, caused a 75% inhibition in mannose incorporation into secreted glycoproteins, but had no effect on the incorporation of [3H]leucine into the secreted glycoproteins. Since deoxymannojirimycin also strongly inhibited mannose incorporation into lipid-linked oligosaccharides in PBS, the decreased amount of radioactivity in the secreted and cellular glycoproteins may reflect the formation of glycoproteins with fewer than normal numbers of oligosaccharide chains, owing to the low levels of oligosaccharide donor. However, in BME medium, there was only slight inhibition of mannose incorporation into lipid-linked saccharides and into cellular and secreted glycoproteins.